Re: [Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-07-22 Thread Stefano Stabellini
On Wed, 22 Jul 2015, Jan Beulich wrote: > >>> On 22.07.15 at 16:49, wrote: > > On Wed, 22 Jul 2015, Jan Beulich wrote: > >> >>> On 21.07.15 at 18:18, wrote: > >> > On Thu, 18 Jun 2015, Jan Beulich wrote: > >> >> --- a/xen/arch/x86/hvm/hvm.c > >> >> +++ b/xen/arch/x86/hvm/hvm.c > >> >> @@ -921,7 +

Re: [Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-07-22 Thread Jan Beulich
>>> On 22.07.15 at 16:49, wrote: > On Wed, 22 Jul 2015, Jan Beulich wrote: >> >>> On 21.07.15 at 18:18, wrote: >> > On Thu, 18 Jun 2015, Jan Beulich wrote: >> >> --- a/xen/arch/x86/hvm/hvm.c >> >> +++ b/xen/arch/x86/hvm/hvm.c >> >> @@ -921,7 +921,7 @@ static void hvm_ioreq_server_disable(str >> >

Re: [Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-07-22 Thread Stefano Stabellini
On Wed, 22 Jul 2015, Jan Beulich wrote: > >>> On 21.07.15 at 18:18, wrote: > > On Thu, 18 Jun 2015, Jan Beulich wrote: > >> --- a/xen/arch/x86/hvm/hvm.c > >> +++ b/xen/arch/x86/hvm/hvm.c > >> @@ -921,7 +921,7 @@ static void hvm_ioreq_server_disable(str > >> > >> static int hvm_ioreq_server_init

Re: [Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-07-22 Thread Jan Beulich
>>> On 21.07.15 at 18:18, wrote: > On Thu, 18 Jun 2015, Jan Beulich wrote: >> --- a/xen/arch/x86/hvm/hvm.c >> +++ b/xen/arch/x86/hvm/hvm.c >> @@ -921,7 +921,7 @@ static void hvm_ioreq_server_disable(str >> >> static int hvm_ioreq_server_init(struct hvm_ioreq_server *s, struct domain >> *d, >>

Re: [Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-07-21 Thread Stefano Stabellini
On Thu, 18 Jun 2015, Jan Beulich wrote: > The number of slots per page being 511 (i.e. not a power of two) means > that the (32-bit) read and write indexes going beyond 2^32 will likely > disturb operation. Extend I/O req server creation so the caller can > indicate that it is using suitable atomic

Re: [Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-06-18 Thread Andrew Cooper
On 18/06/15 14:14, Jan Beulich wrote: > The number of slots per page being 511 (i.e. not a power of two) means > that the (32-bit) read and write indexes going beyond 2^32 will likely > disturb operation. Extend I/O req server creation so the caller can > indicate that it is using suitable atomic a

[Xen-devel] [PATCH v2] x86/HVM: avoid pointer wraparound in bufioreq handling

2015-06-18 Thread Jan Beulich
The number of slots per page being 511 (i.e. not a power of two) means that the (32-bit) read and write indexes going beyond 2^32 will likely disturb operation. Extend I/O req server creation so the caller can indicate that it is using suitable atomic accesses where needed (not all accesses to the